After designing your jobs, verify that they function as expected.
Execute the jobs in SAP Data Services Designer to examine the content of the target tables.
Once you have created your project, jobs, and data flows, run the job in Data Services to move data from the source to the target.
Immediate and Scheduled Jobs
You can run jobs in the following two ways:
- Immediate Jobs
Data Services runs batch and real-time jobs immediately from within the Designer. Both the Designer and the designated job server must be active. Use immediate jobs primarily during the development cycle.
- Scheduled Jobs
You can schedule batch jobs using the Data Services Management Console or a third-party scheduler. The job server must be running for the scheduled job to execute.
Note
Manage Execution Options
Before running a job, you can modify the following execution options:
- Modify job properties to apply the same settings to every execution.
- Modify execution properties to apply settings to a specific execution only.
Shared Execution Properties
| Option | Description |
|---|---|
| Print all trace messages | Record all trace messages in the log. |
| Collect statistics for optimization | Collect statistics to help the Data Services optimizer select the best in-memory or pageable cache type. |
| Collect statistics for monitoring | Display cache statistics in the Administrator Performance Monitor. |
| Use collected statistics | Apply the cache statistics collected during a previous job execution. |
Execution Properties Not Available as Job Properties
| Option | Description |
|---|---|
| System configuration | Specify the system configuration for the job. A system configuration defines a set of datastore configurations, which in turn define the datastore connections. |
| Job server or server group | Select the job server or server group to execute the job. |
Data Services can record failed request information in a log file. This feature is disabled by default to maintain performance. You must enable it in the Management Console Administrator.
Check for Errors
If an execution fails, the job log provides detailed error descriptions.
During execution, Data Services generates three log files accessible via the Monitor tab in the project area. By default, these logs appear automatically in the workspace when a job runs.
Use the Trace, Monitor, and Error icons to view the logs generated during execution.
If the Error icon (the cross) is colored rather than grayed out, errors occurred during execution.
The trace log may indicate a successful completion even if errors exist. This occurs if the data flow is configured to capture specific issues, such as data conversion errors during file access or loading.
Work Flows
After creating projects, batch jobs, and data flows for ETL tasks, consider the benefits of using work flows.
While optional, work flows provide significant advantages in job design.
You can nest work flows within a job and place data flows inside them rather than directly in the job structure.
The following analogies help explain the role of work flows:
For programmers, work flows are similar to subroutines. The job acts as the main program that calls work flows. Work flows perform most job functions but cannot run independently; they must be called by a job or another work flow.
For non-programmers, think of work flows like parentheses in math. Parentheses control the order of operations. Similarly, work flows control the execution order of scripts and data flows within a job.
For basic jobs, work flows may not be necessary. However, as complexity increases with multiple data flows and scripts, work flows make components modular and flexible without impacting performance. They are also reusable across different jobs.
While data flows are reusable, you can group dependent data flows and scripts into a work flow to reuse the entire sequence. Since scripts themselves are not reusable objects, wrapping them in a work flow saves significant design time.
Summary
In this lesson you have learned how to build a data flow in SAP Data Services Designer:
- Begin a data flow by adding source objects from datastores or flat files.
- Add a target object, such as a table or file.
- Use a template table as target if you need to create a new table.
- Use the Query transform to filter, join, and map data from source to target.
- Execute the job to run the data flow and move your data to the target.